Solar and Stellar Astrophysics 13 JAN, 2015

Sensitivity of astrophysical reaction rates to nuclear uncertainties

Sensitivity of astrophysical reaction rates to nuclear uncertainties

Sensitivities of nuclear reaction rates to a variation of nuclear properties are studied. Target nuclei range from proton- to neutron-dripline for 10<=Z<=83. Reactions considered are nucleon- and alpha-induced reactions mediated by the strong interaction. The contribution of reactions proceeding on the target ground state to the total stellar rate is also given. General dependences on various input quantities are discussed. Additionally, sensitivities of laboratory cross sections of nucleon-, alpha-, and gamma-induced reactions are shown, allowing to estimate the impact of cross section measurements. Finally, recommended procedures to explore and improve reaction rate uncertainties using the present sensitivity data are outlined.

💡 Research Summary

The paper presents a comprehensive sensitivity analysis of astrophysical nuclear reaction rates with respect to uncertainties in key nuclear input quantities. The authors focus on a broad range of target nuclei (Z = 10–83), covering isotopes from the proton‑drip line to the neutron‑drip line, and examine nucleon‑ and alpha‑induced reactions that proceed via the strong interaction. Six reaction channels are considered for each projectile: (p,γ), (p,n), (p,α), (n,γ), (n,p), (n,α), (α,γ), (α,n), and (α,p). The study employs the Hauser‑Feshbach statistical model as the theoretical framework for rate calculations, which is the standard approach in large‑scale nucleosynthesis networks.

Three principal nuclear‑physics inputs are varied independently: nuclear level density (NLD), optical model potentials (OMP) for nucleons and alphas, and the γ‑strength function (γSF). Each parameter is perturbed by ±30 % around a baseline value, and the resulting fractional change in the reaction rate is quantified by a sensitivity coefficient S = (ΔR/R)/(ΔX/X). By scanning the entire Z range, the authors map out where each input dominates the uncertainty budget.

Key findings include:

  1. NLD Sensitivity – NLD exerts the strongest influence on neutron‑capture and alpha‑emission reactions, especially for nuclei near the neutron‑drip line where the level density is low. A 30 % change in NLD can increase or decrease the corresponding rate by a factor of two or more. The effect is most pronounced at the low energies relevant to stellar temperatures (kT ≈ 30 keV).

  2. OMP Sensitivity – Optical potentials dominate the uncertainties of proton‑ and alpha‑induced reactions. At sub‑MeV projectile energies (the regime of most astrophysical sites), a ±30 % variation in the OMP leads to 40–60 % changes in the calculated rates. The impact grows with increasing nuclear charge Z, reflecting the heightened sensitivity of the Coulomb barrier to the details of the potential.

  3. γSF Sensitivity – The γ‑strength function is the principal source of uncertainty for γ‑induced reactions such as (γ,n), (γ,p), and (γ,α). In the high‑energy γ‑ray regime that drives the p‑process, variations in γSF can modify rates by more than 70 %. At lower γ energies (≤ 1 MeV) the effect is modest, but it becomes dominant once the photon energy exceeds ~5 MeV.

  4. Ground‑State Contribution – The authors compute the fraction of the stellar rate that originates from reactions on the target ground state. At low stellar temperatures (T ≲ 0.1 GK) the ground‑state contribution exceeds 80 % for most nuclei, justifying the use of laboratory cross sections. However, at higher temperatures (T ≳ 2 GK) excited‑state populations dominate, reducing the ground‑state share to below 30 %. Consequently, experimental data obtained solely on ground‑state targets become insufficient for accurate stellar rate predictions in hot environments.

  5. Laboratory Cross‑Section Sensitivities – By restricting the analysis to the typical experimental energy window (0.1–10 MeV), the authors show that the same nuclear inputs affect measured cross sections differently from their impact on stellar rates. For alpha‑capture experiments, OMP uncertainties dominate (≈ 30 % effect), whereas γSF contributes only ≈ 10 %. Conversely, for γ‑induced measurements, γSF governs the uncertainty (≈ 70 %). This distinction provides a clear prescription for prioritizing experimental efforts: improve OMPs for charged‑particle capture studies and refine γSFs for photodisintegration work.

  6. Uncertainty Propagation Framework – The paper outlines a practical method to propagate nuclear‑physics uncertainties into reaction‑rate uncertainties. Using the calculated sensitivity coefficients, one can perform Monte‑Carlo sampling of the input parameters, generating a probability distribution for each rate. This approach yields statistically robust error bars that can be directly incorporated into nucleosynthesis network calculations.

  7. Recommendations for Future Work – The authors propose a systematic workflow: (i) identify reactions with the largest sensitivity coefficients for a given astrophysical scenario; (ii) target those reactions for high‑precision measurements or refined theoretical modeling; (iii) update the input libraries and recompute sensitivity maps; and (iv) iterate until the overall network uncertainty meets the scientific goals (e.g., reproducing observed abundance patterns within observational errors).

Overall, the study provides a valuable database of sensitivity coefficients that can be used by both experimentalists and modelers. By quantifying which nuclear properties most strongly affect each reaction under realistic stellar conditions, the work guides the allocation of limited resources toward the most impactful measurements. The methodology also offers a transparent way to assess how improvements in nuclear data translate into reduced uncertainties in astrophysical predictions, thereby strengthening the link between nuclear physics experiments and the understanding of stellar nucleosynthesis processes such as the s‑, r‑, p‑, and α‑processes.